It has been estimated that as many as one-half of asthmatic patients do not respond to treatment with [unreadable]2-agonists, glucocorticoids, or leukotriene antagonists. The principle hypothesis of this proposal is that genetics can be used to predict whether an individual with asthma will have a beneficial vs. a poor response to specific asthma therapies. In support of this hypothesis, our laboratories have successfully used a genotype to phenotype candidate gene approach to identify genetic variants that are associated with treatment-specific responses to all three major classes of asthma therapeutics. Despite this success, no variation at the single SNP or haplotypic level in any single gene has been discovered that can explain a large enough proportion of the phenotypic variance to be clinically predictive for any asthma drug. Thus, the major goal of this proposal is to identify a set of robust and clinically significant genetic variants that can predict therapeutic response to asthma drugs. Our goal is to replace the current "trial-and-error" paradigm of asthma treatment with an inexpensive, rapid, and reliable pharmacogenetic test of non-response, to inhaled steroids or [unreadable]2-agonists. As such, we have structured pur Specific Aims as follows: 1. We will choose candidate genes in the [unreadable]2-adrenergic and glucocorticoid pathways, identify genetic variants in these genes, genotype individuals in clinical drug trials at these loci, and perform association testing to determine which genetic variants are associated with an altered therapeutic response. The genes with positive association results that are replicated in multiple study populations will be comprehensively resequenced for functional assessment. 2. Robust and replicated pharmacogenetic associations will be explored at the molecular, cellular, and integrative level to establish functional variants and pharmacogenetic mechanisms. 3. We will develop and validate a predictive test of asthma therapeutic response to [unreadable]2-adrenergic and glucocorticoid drugs. Additional aims relate to our collaboration with PharmGKB and the PGRN and helping investigators interested in pharmacogenetics. DNA samples and matching phenotypic data have been collected from 3698 asthmatics enrolled in 12 well-designed asthma drug trials. This allows us to replicate association findings and validate our pharmacogenetic test in multiple populations. If successful, our research should directly benefit patient care by lowering morbidity and mortality and substantially decreasing health care costs.